WO2022263202A1 - Procédé et système de détermination de valeur cible de paramètre mesuré par un dispositif portable - Google Patents

Procédé et système de détermination de valeur cible de paramètre mesuré par un dispositif portable Download PDF

Info

Publication number
WO2022263202A1
WO2022263202A1 PCT/EP2022/065135 EP2022065135W WO2022263202A1 WO 2022263202 A1 WO2022263202 A1 WO 2022263202A1 EP 2022065135 W EP2022065135 W EP 2022065135W WO 2022263202 A1 WO2022263202 A1 WO 2022263202A1
Authority
WO
WIPO (PCT)
Prior art keywords
time
time window
parameter
target value
value
Prior art date
Application number
PCT/EP2022/065135
Other languages
English (en)
Inventor
Aurélie LE CAIN
Bruno Amir
Original Assignee
Essilor International
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Essilor International filed Critical Essilor International
Priority to CN202280033419.XA priority Critical patent/CN117337474A/zh
Publication of WO2022263202A1 publication Critical patent/WO2022263202A1/fr

Links

Classifications

    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
    • A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
    • A61B5/1118—Determining activity level
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B2560/00—Constructional details of operational features of apparatus; Accessories for medical measuring apparatus
    • A61B2560/02—Operational features
    • A61B2560/0242—Operational features adapted to measure environmental factors, e.g. temperature, pollution
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
    • A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
    • A61B5/112—Gait analysis
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/48—Other medical applications
    • A61B5/4866—Evaluating metabolism
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/68—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient
    • A61B5/6887—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient mounted on external non-worn devices, e.g. non-medical devices
    • A61B5/6898—Portable consumer electronic devices, e.g. music players, telephones, tablet computers
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7235—Details of waveform analysis
    • A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H20/00—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance
    • G16H20/30—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to physical therapies or activities, e.g. physiotherapy, acupressure or exercising
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H80/00—ICT specially adapted for facilitating communication between medical practitioners or patients, e.g. for collaborative diagnosis, therapy or health monitoring

Definitions

  • the invention relates to the field of determining a target value to be reached for a parameter.
  • the method and system described hereinafter are particularly suitable for determining a target number of steps or distance to be travelled for an individual practicing a physical activity and wearing a portable pedometer-type device.
  • a moving-window percentile rank algorithm consists of measuring an individual's physical activity over several days, classifying these days according to the number of steps and then calculating a new goal based on a percentile criterion.
  • the application can thus suggest to the individual to reach, for the following day, a number of steps corresponding to the 60th percentile.
  • certain algorithms propose to calculate a linear moving weighted average.
  • the advantage of such a solution is to attribute a weighting coefficient to each measurement day and therefore to be able to adapt the objective according to the user's daily life.
  • the calculation of the linear moving weighted average does not take a change in an individual's daily behavior into account.
  • this solution does not consider the failure or success of the user to achieve the goals set on the previous days.
  • the present invention seeks to improve the situation.
  • the success factor allows to consider the capacity of the parameter, on one or more preceding time windows, to reach the target value.
  • the retrieving of data related to one or more preceding time windows selected according to the appropriate time scale allows to use only relevant data and in particular to take advantage of a possible periodic behavior in the variations of the parameter value.
  • the value of the parameter is for instance measured by a portable device adapted to be worn by an individual.
  • Such a portable device can be sensitive to motion.
  • the portable device is a pedometer-type device.
  • the parameter quantifies the intensity or quality of an individual's physical activity.
  • the parameter relates for instance to a number of steps, a distance traveled by an individual, a number of calories burned or the duration of the physical activity.
  • the portable device sensitive to motion can also be an inertial measurement device configured to measure motion data of a mobile, for example acceleration or angular velocity.
  • the parameter is therefore measured by a portable device and allows to characterize, for the individual wearing the portable device, his or her physical activity, lifestyle or, more generally, his or her behavior to maintain his or her health.
  • the determined target value then constitutes an objective or goal for the individual
  • the appropriate time scale allows for the individual's habits to be considered.
  • the success factor then corresponds to the individual's past successes or failures in achieving his or her objectives or goals.
  • the success factor is calculated based on a profile of the individual depending on physiological characteristics or data relating to a regular physical activity or health of the individual.
  • the selection of the appropriate time scale depends on the variations in the parameter value over one or more preceding time windows.
  • selecting an appropriate time scale only relevant data are retained. Indeed, the variations of the parameter value can have a periodic behavior. Similarly, the ability of the parameter to reach the target value can also be fluctuating. This is especially true when the parameter quantifies an individual's behavior, such as physical activity.
  • each time scale of the at least two predetermined time scales relates to a period of time.
  • the selection of the appropriate time scale comprises:
  • the stability score assigned to a time scale depends for example on a ratio of pair of two consecutive preceding time windows of the selected data for which the calculated difference is lower than or equal to a predetermined threshold.
  • the selection of the appropriate time scale can be based on the detection of a point event that changes the behavior of the parameter.
  • the parameter allows to quantify the intensity or the quality of the physical activity of an individual
  • the comparison of the value of the parameter from one preceding time window to the next allows to detect a noticeable change in the daily life of the individual when this change has an impact on the physical activity of the individual.
  • each time scale of the at least two predetermined time scales relates to a period of time.
  • the selection of the appropriate time scale comprises: for each time scale of the at least two predetermined time scales: • selecting data relating to one or more preceding time windows spaced from the upcoming time window by a multiple of the period of time to which the time scale relates;
  • the regularity score assigned to a time scale depends on a ratio of preceding time windows of the selected data for which the calculated difference is lower than or equal to a predetermined threshold.
  • the analysis of the variations in the parameter value with respect to the target value determined for one or more preceding time windows allows to detect a change in the performance of the parameter.
  • this embodiment makes it possible to detect a temporary inability to achieve the objectives or goals set by the method, for example during a convalescence.
  • each time scale of the at least two predetermined time scales relates to a period of time
  • the retrieved data relate to one or more preceding time windows spaced from the upcoming time window by a multiple of the period of time to which the selected time scale relates.
  • the determination of the target value for the upcoming time window comprises weighting of the value reached by the parameter over each preceding time window of the retrieved data so that the weighting coefficient of the value reached by the parameter over a given time window is greater than or equal to the weighting coefficient of the value reached by the parameter over a time window prior to the given time window.
  • the target value is determined according to a linearly weighted moving average calculation.
  • the method relies on past data, namely the value reached by the parameter over one or more preceding time windows as well as the target value determined for each of these preceding time windows.
  • the weighting allows to take more recent data than older data.
  • the determined target value is therefore more accurate and realistic.
  • the method further comprises: - generating a time series including values of the parameter, the time series being updated with each new value reached by the parameter over a given time window;
  • the target value for the upcoming time window is thus determined using the at least one predictive model.
  • the time series is segmented into a plurality of groups of time windows according to a criterion related to the respective time components of the time windows and a predictive model is generated by learning for each group of time windows.
  • the determination of the target value for the upcoming time window comprises:
  • the time series can be segmented into four groups so that each group corresponds to a season. This segmentation allows to individualize the data and to take potential drastic changes in the individual's daily life into account.
  • the present invention also relates to a computer program comprising instructions for implementing the previous method, when the instructions are executed by at least one processor.
  • the present invention concerns a system for determining a target value to be reached for a parameter over an upcoming time window comprising:
  • a communication module arranged to receive a set of data relating to a plurality of preceding time windows
  • a storage unit arranged to store the received set of data, the data relating to a preceding time window including the value reached by the parameter over the preceding time window and the associated target value;
  • a processing unit configured to:
  • the system further comprises a portable device adapted to be worn by an individual and arranged to measure the value of the parameter and to transmit the value to the communication module.
  • FIG.1 illustrates a system for determining a target value to be reached for a parameter according to the invention
  • FIG. 2 illustrates a method for determining a target value to be reached for a parameter according to the invention
  • FIG. 3 illustrates a step of selecting an appropriate time scale of the method illustrated in FIG. 2.
  • FIG. 1 illustrates a system 1 for determining a target value to be reached for a parameter.
  • the parameter can be any parameter whose value varies over time and can be controlled, in particular by an individual.
  • the parameter can be used to quantify an aspect of an individual's daily life such as his or her health or physical activity.
  • the system 1 is configured to measure the value of the parameter over time and to determine, at least based on the measurements, a target value that takes the form of a goal for the individual.
  • the system 1 comprises a portable device 3, hereinafter device 3, and a server 5.
  • the device 3 and the server 5 can form a single device or apparatus capable of implementing the functionalities not only of the device 3 but also of the server 5.
  • FIG. 1 simply relates to a particular embodiment in which certain functionalities are deported to a remote entity, namely the server 5.
  • the device 3 is arranged to measure the value of the parameter.
  • the device 3 measures the value of the parameter for example in real time.
  • the device 3 measures the value of the parameter at regular time intervals.
  • the device 3 illustrated in FIG. 1 is a portable device.
  • the portable device 3 is adapted to be worn by an individual. More generally, in the context of the invention, the device 3 is configured to measure a parameter relating to an individual's behavior or activity. This individual is, for example, the wearer of the portable device 3.
  • the portable device 3 is sensitive to motion.
  • the device 3 is for instance an inertial measurement unit (known under the acronym “IMU”).
  • IMU inertial measurement unit
  • the device 3 is an electronic device configured to measure motion data of a mobile, for example acceleration or angular velocity.
  • the device 3 can also be a pedometer-type device.
  • the device 3 is configured to measure movement data relating to the walking of an individual.
  • the device 3 can thus be worn by an individual and is configured to measure a number of steps or a distance traveled by the individual.
  • the parameter measured by the device 3 may be any parameter quantifying the intensity or quality of an individual's physical activity. Therefore, the device 3 can also be configured to measure a number of calories burned or the duration of a physical activity.
  • the device 3 can also be configured to measure other parameters.
  • the device 3 includes for instance an ambient light sensor (known under the acronym “ALS”) configured to measure a parameter characterizing the time spent outside by an individual.
  • the device 3 can also include an ultraviolet sensor configured to measure an amount of vitamin D produced by an individual's body.
  • the device 3 can include a time-of-flight (known under the acronym “TOF”) sensor configured to measure a parameter characterizing the time spent near to work by an individual.
  • TOF time-of-flight
  • the measurements collected by the device 3 are intended to be used to determine a target value of the parameter. This target value can thus be a target number of steps or distance to be covered for the individual wearing the device 3.
  • the device 3 comprises a memory 7, a processor 9, a communication module 11 and a screen 13.
  • the memory 7 is configured to store instructions whose implementation by the processor 9 results in the operation of the device 3. Furthermore, the memory 7 is also configured to store measurements collected by the device 3. The memory 7 may be a volatile or non-volatile memory, and the stored measurements may be erased after using these measurements to determine the target value to be reached by the parameter.
  • the communication module 11 is configured to transmit the measurements collected by the device 3 and stored in the memory 7 to the server 5.
  • the communication module 11 is simply configured to route the measurements stored in the memory to a processing unit of that device or apparatus.
  • the communication module 11 may integrate one or more communication sub-modules, for example for radiofrequency communication and be configured for the transmission and reception of radiofrequency signals, according to one or more technologies, such as TDMA, FDMA, OFDMA, CDMA, or one or more radiocommunication standards, such as GSM, EDGE, CDMA, UMTS, FISPA, LTE, LTE-A, WiFi (IEEE 802. 11) and WiMAX (IEEE 802.16), or their variants or evolutions, currently known or further developed, e.g. 5G.
  • technologies such as TDMA, FDMA, OFDMA, CDMA, or one or more radiocommunication standards, such as GSM, EDGE, CDMA, UMTS, FISPA, LTE, LTE-A, WiFi (IEEE 802. 11) and WiMAX (IEEE 802.16), or their variants or evolutions, currently known or further developed, e.g. 5G.
  • the screen 13 is configured to display information that may be relevant to any individual, especially the individual wearing the device 3 in the embodiment in which the device is a portable pedometer-type device sensitive to motion.
  • the screen 13 is configured to display the number of steps or the distance traveled. This information can be displayed in real time or at regular time intervals, for example at the end of the day.
  • the screen 13 can be configured to display the target value to be reached by the parameter.
  • the screen 13 may be a touch screen or a Human-Machine Interface configured to allow the individual wearing the device 3 to input information necessary for determining the target value to be reached by the parameter.
  • the information can relate to a profile of the individual including physiological characteristics or data relating to a regular physical activity or health of the individual.
  • the server 5 is configured to determine the target value to be reached according to at least the measurements collected by the device 3.
  • the individual wearing the device 3 can also enter information via the screen 13.
  • the information can also be transmitted by the communication module 13 to the server 5 for the determination of the target value.
  • the server 5 comprises a communication module 15, a storage unit 17 and a processing unit 19.
  • the communication module 15 is configured to receive measurements, and possibly profile information of the individual wearing the device 3, from the device 3 and more precisely from the communication module 11.
  • the communication module 15 can form with the communication module 11 means for routing the measurements of the value of the parameter to the processing unit 19.
  • the communication module 15 of the server 5 can be similar to the communication module 11 of the device 3.
  • the storage unit 17 is arranged to store a set of data.
  • the set of data stored in the storage unit 17 includes at least, for one or more time windows, the value reached by the parameter over each of these time windows and the target value determined for each time window.
  • a time window is a fixed time interval that can be expressed in a time unit.
  • a time window is for example a day as clearly defined in a calendar.
  • any other segmentation of time is possible to determine a plurality of time windows of fixed duration.
  • the storage unit 17 can be further configured to store additional information, such as information about the individual's profile and including physiological characteristics or data relating to the regular physical activity or health of the individual.
  • the storage unit 17 can also be configured to store a weighting coefficient associated with each time window and more specifically with each value reached by the parameter over a time window. Indeed, it may be relevant, in order to determine the target value to be reached by the parameter, to weight the values reached by the parameter over previous time windows. Typically, the weighting coefficients are determined so that the weighting coefficient of a given time window is greater than the weighting coefficient of a time window prior to the given time window. Such an embodiment is particularly suitable when the target value is determined according to a linearly weighted moving average calculation.
  • the processing unit 19 is configured to determine the target value to be reached by the parameter over an upcoming time window based on data stored in the storage unit 17.
  • the processing unit 19 comprises a memory 21 and a processor 23.
  • the memory 21 is configured to store instructions whose implementation by the processor 23 results in the operation of the processing unit 19.
  • the operation of the processing unit 19 results in the determination of the target value to be reached by the parameter for the upcoming time window. This target value can then be stored in the storage unit 17 in association with the upcoming time window.
  • the method described hereinafter aims at determining a target value to be reached by a parameter over an upcoming time window.
  • This target value is determined not only on the basis of measurements of the value of the parameter over one or more previous time windows but also on the basis of information such as success or failure in reaching the target value over one or more previous time windows.
  • the parameter is used to characterize the behavior or lifestyle of an individual and quantifies physical activity, diet or health of the individual. The target value determined for the upcoming time window thus constitutes a goal for the individual.
  • the parameter relates to a number of steps or a distance traveled by an individual measured by a portable pedometer-type device sensitive to motion, such as the device 3.
  • the parameter can also be any parameter quantifying the intensity or quality of an individual's physical activity.
  • the parameter can thus relate to a number of calories burned or the duration of a physical activity.
  • the parameter can also relate to the time spent outside by an individual or an amount of vitamin D produced by an individual's body.
  • a set of data is provided.
  • the set of data relates to preceding time windows and includes, for each preceding time window, the value reached by the parameter over the preceding time window and the target value associated with the preceding time window.
  • the set of data is for instance stored in the storage unit 17 and is provided to the processing unit 19 for the determination of the target value to be reached by the parameter over the upcoming time window.
  • the set of data may include additional data such as information relating to the profile of the individual and including physiological characteristics or data relating to the regular physical activity or health of the individual.
  • the set of data also includes a weighting coefficient for each time window and more specifically for each value reached by the parameter over a time window.
  • an appropriate time scale is selected. More exactly, the appropriate time scale is selected among at least two predetermined time scales.
  • the memory 21 of the processing unit 19 stores for instance a plurality of possible time scales among which the appropriate time scale can be selected.
  • a time scale is characterized by a period of time. The time scale makes it possible to select, among the preceding time windows, the relevant time windows, i.e. the time windows associated with data, for instance within the storage unit 17, which have to be used to determine the target value.
  • the method is specifically applied to the embodiment wherein the parameter relates to a number of steps or a distance traveled by an individual measured by the portable pedometer-type device 3 sensitive to motion.
  • the value of the parameter depends on the physical activity of the individual. However, the physical activity depends on the habits of the individual and more generally on the organization of his or her daily life. Therefore, it is necessary to select an appropriate time scale to use the relevant data.
  • the time window corresponds to a day as defined in a calendar.
  • the time scale corresponds for example to a day, in which case the relevant data relate to the consecutive day or days preceding the upcoming day.
  • Another possible time scale is a week.
  • the relevant data are not the data of the previous day but that of the same day of the previous week. Indeed, if the individual practices a physical activity every week only on Tuesday, the target number of steps or the target distance to be traveled for a Tuesday will not be determined according to the data relating to the day before or to the weekend. The data to be considered will indeed be the data relating to the Tuesday of the previous week(s).
  • time scales are possible, for example a time scale corresponding to one month.
  • the processing unit 19 tests one or more time scales to determine the most suitable for determining the target value.
  • the principle is to analyze the variations in the parameter value for one or more preceding time windows.
  • the memory 21 of the processing unit 19 stores a plurality of possible time scales, for instance N time scales, where N is an integer greater than or equal to 2.
  • a local variable is initialized. This local variable scans the set of integers from 1 to N and allows to characterize each predetermined time scale.
  • each time scale corresponds to a period of time. The following sub-steps are applied to each time scale, and more particularly to the associated period of time.
  • the processing unit 19 selects data relating to one or more preceding time windows spaced from the upcoming time window by a multiple of the period of time to which the time scale relates are selected.
  • the data are typically retrieved from the storage unit 17.
  • the processing unit 19 selects at least the data corresponding to the same day of the previous week, thus the day spaced one week from the upcoming time window, i.e. the upcoming day.
  • the processing unit 19 can also select data for the same day of several previous weeks, for example the day two weeks away from the upcoming day and the day three weeks away from the upcoming day.
  • the processing unit 19 can select data relating to the previous five days, since each of these days is separated from the upcoming day by a multiple of one day.
  • the processing unit 19 calculates, for each preceding time window of the selected data, a difference between the value reached by the parameter over the preceding time window and the target value associated with the considered preceding time window.
  • the storage unit 17 stores data relating to several preceding time windows. For each preceding time window, the storage unit 17 stores both the value reached by the parameter over that preceding time window and the target value determined for that preceding time window.
  • the difference is for example the absolute value of the difference between the reached value and the target value.
  • the processing unit 19 selects the data corresponding to the last five time windows, i.e. the last five days, the processing unit 19 calculates five differences, i.e. the difference between the value reached by the parameter and the target value for each of these last five days.
  • the processing unit 19 assigns to the tested time scale a regularity score characterizing the one or more calculated differences.
  • the regularity score assigned to the time scale depends for instance on a ratio of preceding time windows of the selected data for which the calculated difference is lower than or equal to a predetermined threshold.
  • the processing unit 19 determines, according to the value of the local variable, whether all the time scales have been tested. Otherwise, the following time scale is selected to be tested according to the sub-steps S21 , S22 and S23 as explained previously.
  • the processing unit 19 selects the appropriate time scale based on the respective regularity scores of the time scales of the at least two predetermined time scales.
  • the appropriate time scale is for instance the one with the highest score.
  • the memory 21 of the processing unit 19 can store a plurality of possible time scales, for instance N time scales, where N is an integer greater than or equal to 2. At the beginning of step S2, a local variable is initialized.
  • This local variable scans the set of integers from 1 to N and allows to characterize each predetermined time scale.
  • each time scale corresponds to a period of time.
  • the following sub-steps are applied to each time scale, and more particularly to the associated period of time.
  • the processing unit 19 selects data relating to preceding time windows spaced from the upcoming time window by a multiple of the period of time to which the time scale relates are selected.
  • the data are typically retrieved from the storage unit 17.
  • the processing unit 19 calculates, for each pair of two consecutives preceding time window of the selected data, a difference between the values reached by the parameter respectively over each preceding time window of the pair of two consecutives preceding time windows.
  • two preceding time windows of a pair are temporally consecutive.
  • two preceding time windows are consecutive if they follow each other. For example, if the time scale corresponds to one week, two consecutive preceding time windows are separated by exactly one week.
  • the difference is for example the absolute value of the difference between the values reached by the parameter respectively over each preceding time window of the pair of two consecutives preceding time windows.
  • the processing unit 19 calculates three differences.
  • the processing unit 19 calculates a first difference between the value reached by the parameter over the first day and the value reached by the parameter over the second day, a second difference between the value reached by the parameter over the second day and the value reached by the parameter over the third day and a third difference between the value reached by the parameter over the third day and the value reached by the parameter over the fourth day.
  • the processing unit 19 assigns to the tested time scale a stability score characterizing the one or more calculated differences.
  • the stability score assigned to the time scale depends for instance on a ratio of pair of two consecutive preceding time windows of the selected data for which the calculated difference is lower than or equal to a predetermined threshold.
  • the predetermined ratio used to determine the stability can be equal to or different from the predetermined threshold used to determine the regularity ratio.
  • the processing unit 19 has thus calculated the difference between the values reached by the parameter respectively over each preceding time window of each of the four pairs of consecutives days. Furthermore, if the parameter corresponds to the traveled distance, the predetermined threshold can be set to 2 kilometers. The stability score can then correspond to the proportion of these four pairs of consecutive days for which there is less than 2 kilometers difference between the distance traveled by the individual over the first day of a pair and the second day of the same pair.
  • the processing unit 19 determines, according to the value of the local variable, whether all the time scales have been tested. Otherwise, the following time scale is selected to be tested according to the sub-steps S21 , S22 and S23 as explained previously. Finally, in the sub-step S25, the processing unit 19 selects the appropriate time scale based on the respective stability scores of the time scales of the at least two predetermined time scales.
  • the appropriate time scale is for instance the one with the highest score.
  • the two embodiments previously described with reference to FIG. 3 can also be combined. For example, it is possible to calculate, for each time scale, both the regularity score and the stability score and then to compare the scores of the time scales to select the appropriate time scale. Another possible combination is also to choose one or the other of the embodiments according to the time scale selected for the time window immediately preceding the upcoming time window.
  • the processing unit 19 retrieves data relating to one or more preceding time windows based on the selected appropriate time scale.
  • the retrieved data relate to one or more preceding time windows spaced from the upcoming time window by a multiple of the period of time to which the selected appropriate time scale relates. It must be noted that the retrieved data are not necessarily exactly the ones selected in sub-step S21 for testing the finally selected time scale. If the selected time scale corresponds to a week and if three days were selected in sub-step S21 , the processing unit 19 can, in step S3, select six days, i.e. the same day as the upcoming day, but for the last six weeks. In a step S4, the processing unit 19 calculates a success factor based on the comparison between the value reached by the parameter over one or more preceding time windows and the associated target value.
  • one of the advantages of the invention is to take the past success or failure to reach the target value into account.
  • the success factor makes it possible to take into account the individual's ability to succeed, the day before or the days before, in reaching the set goal, i.e. the determined target value.
  • This success factor thus makes it possible to consider the individual's recent performance and his or her ability to achieve his or her objectives.
  • This success factor typically takes the form of a multiplicative coefficient greater than or equal to 1 when the individual has globally succeeded, over one or more days preceding the day to come, in reaching the set goal.
  • this multiplicative coefficient is less than 1 when the individual has globally failed, over one or more days preceding the day to come, to reach the set goal.
  • this reasoning can be applied to any parameter, and not only the number of steps or the distance covered by an individual measured by a portable pedometer-type device sensitive to motion.
  • the value of the success factor can be determined as follows: the processing unit 19 selects the last three preceding time windows, so typically the last three days. For each of these last three days, the storage unit 17 stores the value reached by the parameter, for instance the number of steps or the distance traveled by an individual, as well as the target value determined for each of these three days. If the value of the parameter is greater than the associated target value for all three days, then the success factor is equal to 1.1 (or 110%). On the other hand, if the value of the parameter is lower than the associated target value for the three days considered, then the success factor is equal to 0.9 (or 90%). In the other cases, if the value of the parameter is higher than the associated target value only for one or two days, then the success factor is equal to 1 (or 100%).
  • the success factor can be calculated based on a profile of the individual depending on physiological characteristics or data relating to the regular physical activity or health of the individual.
  • the processing unit 19 determines the target value for the upcoming time window based on the success factor and the value reached by the parameter over each preceding time window of the retrieved data.
  • the retrieved data correspond here to the data retrieved during step S3.
  • the target value to be reached by the parameter over an upcoming time window j can be calculated as follows: where:
  • - TV(j) is the target value to be reached by the parameter over the upcoming time window;
  • - k is the number of preceding time window of the retrieved data;
  • - P(j-i) is the value reached by the parameter over the i-th preceding time window
  • the step S5 comprises a preliminary sub-step S51 wherein the value reached by the parameter over each preceding time window of the retrieved data is weighted.
  • the weighting of the value reached by the parameter over a given time window is greater than the weighting of the value reached by the parameter over a time window prior to the given time window.
  • This weighting sub-step S51 is particularly suitable when the target value is determined according to a linearly weighted moving average calculation.
  • the target value to be reached by the parameter over an upcoming time window j can be calculated as follows: where:
  • - TV(j) is the target value to be reached by the parameter over the upcoming time window;
  • - k is the number of preceding time window of the retrieved data;
  • - w(j-7i) is the weighting coefficient of the i-th preceding time window
  • - P(j-7i) is the value reached by the parameter over the i-th preceding time window; and - SF is the success factor.
  • the steps described below relate to a particular embodiment in which the target value to be reached by the parameter over the upcoming time window is also determined by learning.
  • the device 3 measures parameter values.
  • the parameter values can be measured in real time or at regular time intervals. These values are measured over one or more time windows and are transmitted by the communication module 11 of the device 3 to the communication module 15 of the server 5 to be stored in the storage unit 17.
  • the device 3 is for example a portable pedometer-type device sensitive to motion and the measured parameter is the number of steps or the distance traveled by the individual wearing the device 3.
  • the measured values can be displayed in real time or at regular time intervals by the screen 13 of the device 3.
  • the measurements collected by the device 3 and transmitted to the server 5 are used to generate the data provided to the processing unit 19 in order to determine the target value to be reached by the parameter over the upcoming time window.
  • this step T1 is not unique to the embodiment in which the target value is determined at least in part by learning and can therefore be implemented independently, if only to generate the data stored in the storage unit 17.
  • a time series is a series of data points indexed or listed in time order.
  • this time series takes the form of a set of parameter values each associated with a time component which characterizes, for example, the instant at which the value was measured.
  • the time series can be updated with each new value reached by the parameter over a given time window.
  • the time series is updated each time the device 3 measures a new value of the parameter and transmits this value to the server 5.
  • the processing unit 19 segments the time series into a plurality of groups of time windows according to a criterion related to the respective time components of the time windows.
  • the temporal components of the time windows allow to order them in time, i.e. to classify them chronologically.
  • the temporal component of a time window allows to know, for example, if this time window is earlier or later than another time window, if the temporal component associated with the latter is known. Therefore, this criterion consists, for instance, in grouping consecutive time windows once they have been classified chronologically. According to an embodiment, each group comprises the same number of time windows.
  • the time series can be segmented into four groups so that each group corresponds to a season. This segmentation allows to individualize the data and to take potential drastic changes in the individual's daily life into account.
  • the processing unit 19 In a step T4, the processing unit 19 generates by learning at least one predictive model of the value of the parameter for a given time window as a function of a time component characterizing the considered time window. It can be supervised or semi-supervised learning.
  • step T3 a predictive model is generated by learning for each group of time windows. Conversely, if the time series is not segmented by the processing unit 19, only one predictive model can be generated in step T4. In a step T5, the processing unit 19 applied the criterion to the time component of the upcoming time window in order to determine the corresponding group of time windows.
  • step T5 determines, using the time component of the upcoming time window, therefore the day to come, to which season this upcoming day belongs.
  • the temporal component thus makes it possible to determine to which group of the time series the upcoming day is closest.
  • step S5 illustrated in FIG. 2 the predictive model associated with said group of time windows is used to determine the target value for the upcoming time window.
  • the predictive model can be used to determine, depending on the time component of the upcoming time window, the target value to be reached by the parameter on the upcoming time window.
  • the predictive model can be used alone to determine the target value but can also be used in combination with the embodiments detailed previously, i.e. with the weighting coefficients for the value reached by the parameter over one or more preceding time windows and with the data relating to the profile of the individual including physiological characteristics or data relating to the regular physical activity or health of the individual.

Abstract

L'invention concerne un procédé de détermination d'une valeur cible pour un paramètre sur une fenêtre temporelle à venir comprenant les étapes suivantes : - fournir (S1) un ensemble de données comprenant, pour chaque fenêtre temporelle précédente, la valeur atteinte par le paramètre et la valeur cible associée ; - sélectionner (S2) une échelle de temps appropriée parmi au moins deux échelles de temps ; - récupérer (S3) des données se rapportant à des fenêtres temporelles précédentes en fonction de l'échelle de temps ; - calculer (S4) un facteur de réussite en fonction de la comparaison entre la valeur atteinte par le paramètre sur une ou plusieurs fenêtres temporelles précédentes et la valeur cible associée ; et - déterminer (S5) la valeur cible en fonction du facteur de réussite et de la valeur atteinte par le paramètre sur chaque fenêtre temporelle précédente des données récupérées. Le paramètre peut être un nombre de pas mesuré par un dispositif de type podomètre portable porté par un individu.
PCT/EP2022/065135 2021-06-14 2022-06-02 Procédé et système de détermination de valeur cible de paramètre mesuré par un dispositif portable WO2022263202A1 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202280033419.XA CN117337474A (zh) 2021-06-14 2022-06-02 用于确定由便携式装置测量的参数的目标值的方法和系统

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
EP21305812.6A EP4105942A1 (fr) 2021-06-14 2021-06-14 Procédé et système de détermination d'une valeur cible d'un paramètre mesuré par un dispositif portatif
EP21305812.6 2021-06-14

Publications (1)

Publication Number Publication Date
WO2022263202A1 true WO2022263202A1 (fr) 2022-12-22

Family

ID=76708172

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/EP2022/065135 WO2022263202A1 (fr) 2021-06-14 2022-06-02 Procédé et système de détermination de valeur cible de paramètre mesuré par un dispositif portable

Country Status (3)

Country Link
EP (1) EP4105942A1 (fr)
CN (1) CN117337474A (fr)
WO (1) WO2022263202A1 (fr)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140135592A1 (en) * 2012-11-13 2014-05-15 Dacadoo Ag Health band
US20160371998A1 (en) * 2015-06-17 2016-12-22 Benecure, Inc. Health-based incentive plans and systems and methods for implementing health-based incentive transactions
US10602964B2 (en) * 2016-08-17 2020-03-31 Koninklijke Philips N.V. Location, activity, and health compliance monitoring using multidimensional context analysis

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140135592A1 (en) * 2012-11-13 2014-05-15 Dacadoo Ag Health band
US20160371998A1 (en) * 2015-06-17 2016-12-22 Benecure, Inc. Health-based incentive plans and systems and methods for implementing health-based incentive transactions
US10602964B2 (en) * 2016-08-17 2020-03-31 Koninklijke Philips N.V. Location, activity, and health compliance monitoring using multidimensional context analysis

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
GRAHAM BOWER: "10 hidden Activity app features that will take your fitness to the next level", 16 April 2019 (2019-04-16), XP055868737, Retrieved from the Internet <URL:https://www.cultofmac.com/616240/hidden-activity-app-features/> [retrieved on 20211202] *

Also Published As

Publication number Publication date
EP4105942A1 (fr) 2022-12-21
CN117337474A (zh) 2024-01-02

Similar Documents

Publication Publication Date Title
US11621903B2 (en) Combining measurements based on beacon data
US9830607B1 (en) Multi-platform overlap estimation
EP2819383B1 (fr) Système de suivi d&#39;activité d&#39;utilisateur et dispositif
US10709933B2 (en) Pose and heart rate energy expenditure for yoga
US10565955B2 (en) Display status adjustment method, display status adjustment device and display device
US9750433B2 (en) Using health monitor data to detect macro and micro habits with a behavioral model
US7327245B2 (en) Sensing and analysis of ambient contextual signals for discriminating between indoor and outdoor locations
US20160081620A1 (en) Method and apparatus for health care
CN110709940B (zh) 用于预测传感器测量质量的方法、系统和介质
US20150269195A1 (en) Model updating apparatus and method
US11715049B2 (en) Information processing device and method
US20230418357A1 (en) Smart Context Subsampling On-Device System
JP2015010927A (ja) 観測値処理装置、観測値処理方法及び観測値処理プログラム
US20180366024A1 (en) Providing suggested behavior modifications for a correlation
CN111513723A (zh) 运动姿态监测方法、调整方法、装置和终端
CN110544132A (zh) 用户常活动位置的确定方法、装置、设备和存储介质
EP4105942A1 (fr) Procédé et système de détermination d&#39;une valeur cible d&#39;un paramètre mesuré par un dispositif portatif
Armah et al. A deep analysis of the precision formula for imbalanced class distribution
JP6830685B1 (ja) リンゴ品質推定プログラム及びシステム
JP2017012277A (ja) 携帯型電子装置、センサ制御システム、センサ制御方法及びセンサ制御プログラム
US20180001140A1 (en) Exercise schedule optimizer system
CN109642827B (zh) 温度推算系统、温度推算方法及记录有温度推算程序的记录媒体
Jabla et al. Balancing Timing and Accuracy Requirements in Human Activity Recognition Mobile Applications
US20220096896A1 (en) Measurement device, measurement method, and recording medium
US20240091592A1 (en) Method and server for providing exercise management solution

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22730929

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE